← Back to community benchmarks
Qwen3.5-VL-122B-A10Bcrk
Performance
1k
tokens
668.2
PP tok/s
50.6
TG tok/s
1533
TTFT (ms)
66.4
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
96 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 668.2 | 50.6 | 66.4 GB | current |
| 4k | 702.1 | 48.4 | 66.7 GB | view |
| 8k | 687.9 | 45.3 | 67.0 GB | view |
| 16k | 648.3 | 39.5 | 67.7 GB | view |
| 32k | 570.6 | 30.4 | 69.1 GB | view |
| 64k | 450.4 | 21.4 | 71.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 50.6 | 1.00× |
| 2× | 49.8 | 0.98× |
| 4× | 71.7 | 1.42× |
| 8× | 139.2 | 2.75× |